The Use and Perception of Electronic Cigarettes andSnus among the U.S. PopulationShu-Hong Zhu*, Anthony Gamst, Madeleine Lee, Sharon Cummins, Lu Yin, Leslie Zoref

Moores Cancer Center, University of California San Diego, La Jolla, California, United States of America

Abstract

Background: E-cigarettes have generated controversy in the tobacco control field similar to that of Swedish snus,which came to the U.S. market six years earlier. Some argue that e-cigarettes have great potential to help smokersquit regular cigarettes while others contend they should be banned for lack of safety and efficacy data. This studyexamined population data from the U.S.Methods: A U.S. population survey with a national probability sample (N=10,041) was conducted (February 24 toMarch 8, 2012, before any major paid advertisement of e-cigarettes appeared on television). Survey respondentswere asked if they had heard about e-cigarettes, where they had heard about them, whether they had used e-cigarettes or snus, how often they used them, and why they used them. Responses were weighted to represent theentire U.S. population.Findings: A high proportion, 75.4%, reported having heard about e-cigarettes. Television ranked as the number onesource of information, followed by “in-person conversation” and “Internet.” About 8.1% had tried e-cigarettes, and1.4% were current users. These rates were twice those of snus (4.3% and 0.8%, respectively). Among currentsmokers, 32.2% had tried e-cigarettes, and 6.3% were current users. Over 80% of current e-cigarette users werenon-daily users. Women were significantly more likely to have tried e-cigarettes than men. Those who had tried e-cigarettes were more likely than those who tried snus to report their products being safer than regular cigarettes(49.9% vs. 10.8%). Almost half (49.5%) of current smokers were susceptible to using e-cigarettes in the future.Conclusions: That e-cigarettes have surpassed snus in adoption rate, even before any promotion by major tobaccocompanies, suggests that the former have tapped into smokers’ intuitive preference for potentially harm-reducingproducts, probably due to the product design. E-cigarette use is likely to increase in the next few years.

Citation: Zhu S-H, Gamst A, Lee M, Cummins S, Yin L, et al. (2013) The Use and Perception of Electronic Cigarettes and Snus among the U.S.Population . PLoS ONE 8(10): e79332. doi:10.1371/journal.pone.0079332

Editor: Alan Blum, University of Alabama, United States of America

Received July 16, 2012; Accepted October 1, 2013; Published October 24, 2013

Copyright: © 2013 Zhu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by a grant from the National Cancer Institute (http://www.cancer.gov), U01 CA154280. The funders had no role in studydesign, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

* E-mail: szhu@ucsd.edu

Introduction

The Electronic Nicotine Delivery System (ENDS), also knownas the e-cigarette, has increasingly attracted the attention ofsmokers and tobacco control workers [1,2]. E-cigarettes werefirst developed in China in 2003 [3]. They came to the U.S.market in 2007 and quickly gained notoriety in many countries,especially those with relatively strong tobacco control programs[4-9]. Unfortunately, scientific information about e-cigarettes islimited. Some argue that e-cigarettes are obviously lessharmful than cigarettes and have great potential to helpsmokers quit [10,11], while others contend that data on safetyare needed before e-cigarettes are promoted or allowed to besold [1]. Anecdotal reports of smokers using e-cigarettes to

help them quit smoking abound [3,7,12-14], but efficacy data inthe form of clinical trials are still limited [15-19].

Insufficient scientific research on the safety and efficacy of e-cigarettes is one reason that the products have attractedcontroversy. Some countries have banned the sale of e-cigarettes [20], although that does not prevent smokers frompurchasing them on the Internet. In the U.S., the Food andDrug Administration (FDA) has attempted to regulate the saleand marketing of e-cigarettes, a move that was struck down bya federal court [21]. Short of FDA regulatory oversight, somestates have tried to pass laws to ban the sale of e-cigarettes intheir own jurisdictions [22] although the availability of e-cigarettes on the Internet makes it difficult to enforce such aban. Meanwhile, the rationale for the ban itself appears to be

PLOS ONE | www.plosone.org 1 October 2013 | Volume 8 | Issue 10 | e79332

chiefly based on predicted potential harm, as empiricalevidence is sparse [11].

This controversy surrounding e-cigarettes is reminiscent ofthe controversy associated with another tobacco product, snus.Snus, a moist smokeless tobacco product popular in Sweden,gained the attention of global tobacco control workers a fewyears before e-cigarettes did [23]. Strong arguments for andagainst snus have been advanced, but the tobacco control fieldremains divided [24-27]. Especially difficult is the debate on thepotential of snus to reduce the harm of tobacco use at thepopulation level [27]. While using snus may be less risky thansmoking cigarettes to the health of the individual, it is not clearthat promoting the use of snus would reduce the total harmassociated with tobacco use at the population level [28,29].Some have argued that promoting the use of any tobaccoproduct supports the tobacco-use norm and, as such, wouldproduce a negative net-effect on tobacco control at thepopulation level [27]. Similar difficulty exists in the currentcontroversy on e-cigarettes.

There is, however, one noticeable difference in the shorthistory of e-cigarettes and snus in the U.S. market. E-cigarettesseem to have achieved notoriety relatively quickly withoutmajor paid advertising [30]. Unlike snus, which has beenpromoted by large tobacco companies in the U.S. [31,32], e-cigarettes had not been promoted by any major tobaccocompany until Lorillard Inc. acquired a major brand Blu-Cigs inApril 2012 [33]. Instead, e-cigarettes appear to have receivedmuch free publicity. No study has carefully documented thelevel of paid advertising versus earned media for e-cigarettes.But a quick web search will show that e-cigarettes havereceived much free coverage. For example, endorsementshave come from some American celebrities and talk showhosts, who tout e-cigarettes’ intuitive appeal and how they canhelp smokers quit cigarettes [34,35]. E-cigarettes haveappeared in popular movies [35,36]. By 2010, web searches forinformation on e-cigarettes in that year had surpassed those forsnus in the U.S. [6] All this suggests that the adoption of e-cigarettes may be significant, and a comparison with theadoption of snus will be informative.

The present study aimed to provide some basic measures onhow much of a foot-hold e-cigarettes had already taken amongthe U.S. population before Lorillard Inc. purchased a well-known e-cigarette brand and started a significant televisionadvertising campaign [33,37]. Using a survey of a probabilitysample of the U.S. population, this study examined theknowledge about e-cigarettes among smokers andnonsmokers. It provided population prevalence measures onever and current use of e-cigarettes and the rate of transitionfrom ever use to current use. The perceived utility of e-cigarettes as a quitting aid or as a potential harm reductionproduct was assessed and compared with that of snus, apotential harm reduction product that aroused similarcontroversy when it came to the U.S. a few years before e-cigarettes. Finally, the proportion of the U.S. population that issusceptible to future e-cigarettes was estimated.

Methods

Ethics StatementThis research was performed in accordance with a human

subjects protocol approved by the University of California, SanDiego’s Institutional Review Board (IRB# 111664).

Data sourceThe data for this study was obtained from a survey

commissioned by the University of California, San Diego andadministered by Knowledge Networks (Menlo Park, CA).Knowledge Networks, which was recently acquired by GfK,recruits a probability sample representative of the U.S.population (KnowledgePanel). The sample was originallyrecruited by random digit dialing (RDD) but an address-basedsampling methodology has been used in recent years [38]. Adetailed description of the sampling methods used to recruit tothe KnowledgePanel has been described elsewhere [39]. Theadvantages and the limitations of using the KnowledgePanelhave also been discussed in many contexts and will not berepeated here [40-42]. In summary, the panel provides anefficient way of accessing a probability sample of the U.S.population, whose representativeness is similar to most otherwell-known population surveys [42-46]. All KnowledgeNetworks surveys are performed online. Knowledge Networksprovides a netbook computer and network access toparticipants, as needed. Many health behavior studies haveused the KnowledgePanel [45,47,48].

The present survey was designed to gather information onsmoking history and cigarette use, perceptions about differenttobacco products and quitting aids, attitudes toward tobaccocontrol efforts, and beliefs and ideation about the process ofquitting smoking. The study over sampled the smokers in theKnowledgePanel so that all the available smokers wereincluded, with a random sub-sample of former smokers andnever smokers from the panel such that the three smoking-status groups were approximately equal in size. A total of15,095 adults (> 18 years of age) were sampled and invited toparticipate in the survey. Of these, 10,041 completed thesurvey, a response rate of 66.5%. This corresponds to asample with 3,111 current smokers, 3,676 former smokers and3,254 never smokers. The survey was conducted betweenFebruary 24th and March 8th, 2012.

MeasurementCigarette smoking behavior was assessed in multiple

questions. Current smokers were defined as those who hadsmoked at least 100 cigarettes in their lifetime and whoanswered the question, “Do you currently smoke cigarettesevery day, some days, or not at all?” with “every day” or “somedays”. Those who had smoked at least 100 cigarettes in theirlifetime and answered “not at all” were classified as formersmokers. Former smokers were further asked, “When did yousmoke your last cigarette?” They were categorized as recentformer smokers if they selected any of the options with a timeframe of 1 year or less, and long-term former smokers if theyanswered “Over 1 year ago”. Nonsmokers were defined asthose who had not smoked 100 cigarettes in their lifetime.

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 2 October 2013 | Volume 8 | Issue 10 | e79332

Current smokers were asked if they had ever tried to quitsmoking, and if they answered yes, were also asked whetherthey had tried to quit in the last 12 months.

Use of snus was assessed by the question, “Have you everused any of the following tobacco products?” for which “Snus(tobacco in a small pouch, like Camel snus or Marlboro snus)”was one of the available options. Those who selected “yes”were defined as ever users of snus and asked the question “Doyou currently use snus every day, some days, or not at all?”Those selected “every day” or “some days” were defined ascurrent users of snus.

Use of e-cigarettes was also assessed in multiple questions.First, respondents were asked if they have ever heard of e-cigarettes: “E-cigarettes (electronic cigarettes) are electronicdevices that deliver nicotine in a vapor and look like cigarettes,but contain no tobacco. Have you ever heard of e-cigarettes?”Those who had heard of e-cigarettes were also asked wherethey had heard about e-cigarettes and were allowed to selectone or more of the following options: “Radio”, “TV”, “Internet”,“In-person conversation”, “Information shared via Facebook,YouTube, or other social network media”, and “Other”.

Additionally, those who had heard of e-cigarettes wereasked: “Have you ever tried an e-cigarette”, and those whoanswered yes were considered ever users. Ever users werealso asked “Have you used e-cigarettes in the last 30 days”,and those who answered yes were considered current users.Current users were asked to provide the number of days (in thelast 30) they had used e-cigarettes.

Ever users of e-cigarettes were asked “Why did you use e-cigarettes?” and instructed to select “Yes” or “No” for each ofthe following options: “Safer than cigarettes”, “Cheaper thancigarettes,” “Easy to use when I can’t smoke,” “To try to quitsmoking cigarettes”, or “Just because.” Since a person couldhave multiple reasons for using any product, the order of theseoptions was randomized by individual respondent to minimizethe order effect in response (e.g., respondent may be morelikely to choose the first option on the list). The “just because”option was included to make clear that the respondent neednot have any particular reason.

Finally, those who had never used e-cigarettes were askedthe question: “How likely are you to try e-cigarettes in thefuture?” This was intended to assess their susceptibility to e-cigarettes, much like the susceptibility measure on uptake ofregular cigarettes [49]. Those who responded that they were“Very likely” or “Somewhat likely” were considered to besusceptible. This is slightly stricter definition than thesusceptibility measure used in the literature for uptake ofcigarettes in that the present definition does not include thosewho responded “somewhat unlikely” [49]. The susceptibilitydefinition here does include those who have tried e-cigarettesbut are not currently using them.

AnalysisAll percentages were weighted by population parameters

based on the most recent U.S. Current Population Survey [46].A survey-specific post-stratification adjustment was used toaccount for any survey non-response, as well as any non-coverage or under- and over-sampling resulting from the

survey-specific sampling design. In this case, this survey hadapproximately equal numbers of current smokers, formersmokers, and never smokers. The adjustment for over-sampling of smokers produced an overall smoking prevalencefor the U.S. of 19.1% based on this survey, which is quite closeto the newest published national estimate of 19.3% based onthe 2010 National Health Interview Survey [50]. All results wereanalyzed by demographic categories (gender, age, educationallevel, ethnic background), as well as by smoking status.Standard errors were calculated and 95% confidence intervalswere computed based on the sampling distribution of thecorresponding summary statistic. Confidence intervals forbinomial proportions were computed using the method ofAgresti and Coull ([51]; see also 52). All calculations were doneusing R 2.12.1 [53].

Results

Figure 1 shows the rate of having “ever used” and “currentlyuse” for e-cigarettes and snus, weighted to the U.S. population.A total of 8.08% reported that they had ever used e-cigarettes,and 1.44% reported currently using e-cigarettes. Thus,approximately 18% of those who have ever used e-cigarettescontinue as current users (1.44/8.08 =17.8%).

The rate of ever used and current use for snus isapproximately half that of e-cigarettes: 4.26% and 0.77%,respectively. The rate of transition from ever users of snus tocurrent users, however, is about the same, 18% (0.77/4.26 =18.1%).

Table 1 presents the usage rates of e-cigarettes and snus bydemographic categories. It also separates out those who usedonly e-cigarettes or snus from those who used both products.The top half of the table shows the “ever use” rates. Womenare more likely to have ever used e-cigarettes than men. Theusage rate is higher among the young and those with lowereducation. Hispanics are less likely to have ever used e-cigarettes than either Whites or Blacks.

The usage rates for snus are significantly lower. The maindifference is in gender: About the same percentage of menhave tried e-cigarettes or snus: 7.17 % (5.33% + 1.84%) for e-cigarettes and 7.10% (5.26% + 1.84%) for snus. However, thepercentage of women having tried e-cigarettes is much higherthan for snus 8.92% (8.33% + 0.59%) versus 1.65% (1.06% +0.59%).

The bottom half of Table 1 shows the rates of current use.The demographic pattern for the “currently use” is similar tothat for “ever used” except that the rates for the former aresignificantly lower across all demographic categories.

Table 2 shows the rates of ever used and current use bysmoking status and by gender. In this table, all users of e-cigarettes are combined into one group regardless of whetherthey use snus or not. The same is done for snus users: Theyare combined into one group of snus users regardless of theire-cigarette use status (thus, dual users are counted in bothcalculations).

Table 2 shows that the difference in usage rates by smokingstatus is large. About 1% of never smokers have ever tried e-cigarettes, while over 32% of current smokers have used e-

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 3 October 2013 | Volume 8 | Issue 10 | e79332

cigarettes. The same is true with ever using snus. Clearly,recent former smokers and current smokers are the most likelyto have tried e-cigarette or snus.

There is also a significant gender difference in ever using e-cigarettes or snus. This gender difference is most clearly seenamong the current smokers: women are more likely to havetried e-cigarettes than men (about 38% vs. 27%). In contrast,men are more likely to have tried snus (about 17% vs. 6%). Itshould be noted, however, that both women and men are morelikely to have tried e-cigarettes than snus.

Table 2 also shows that the current use rates have a similarpattern to the ever use rates. It is recent former smokers andcurrent smokers who are more likely to be current users of e-cigarettes or snus. The gender difference between e-cigaretteand snus use is reversed in this comparison, but the differenceis not statistically significant. The overall rates, however, aresignificantly higher for e-cigarettes than for snus among recentformers smokers and current smokers.

Table 3 shows the frequency of use for e-cigarettes or snusamong those who currently use either of these two types ofproducts. Among e-cigarette users, there is a significantdifference in the use pattern between current smokers andrecent former smokers. Among current smokers, those whoused e-cigarettes were mostly occasional users, only 11.5% ofused e-cigarettes daily. Among recent former smokers,however, 45.7% used them on a daily basis, a statisticallysignificant difference (p<0.05).

Most snus users used snus on a non-daily basis. Among thelong-term former smokers, however, most used snus on a dailybasis.

Table 4 shows the reasons of use, reported by the ever e-cigarette users and ever snus users. It is useful to take a lookat those who used both products, first (the last two columns ofthe table). The most common reason for having tried e-cigarettes or snus is “just because:” 72.3% for e-cigarettes and82.1% for snus. For e-cigarettes, the second most commonreason given is “to try to quit smoking cigarettes,” followed by“safer than cigarettes” and “easy to use when I can’t smoke.”For snus, the second most common reason is “easy to usewhen I can’t smoke,” “followed by to try to quit smokingcigarettes.” Overall, the dual users are significantly more likelyto report use of e-cigarettes than snus to quit smoking, 56.9%vs. 30.1%. Dual users are more likely to report the belief that e-cigarettes are safer than cigarettes, 58.2% vs. 26.2% for snus.

The patterns for those using either e-cigarettes only or snusonly are presented in columns 1 and 2, and they are similar tothat of dual users. E-cigarette users are about twice as likely assnus users to report using the product “to try to quit smoking”(54.9% vs. 26.3%). The former are also significantly more likelyto believe e-cigarettes are safer than regular cigarettes thanthe latter are to believe snus is safer than cigarettes (49.9% vs.10.8%).

Figure 2 shows that those who are currently using e-cigarettes are significantly more likely to have tried to quitsmoking in the last 12 months than those who are currently notusing e-cigarettes. The former are also more likely to have

Figure 1. The rates of ever use and current use of E-Cigarettes and Snus. doi: 10.1371/journal.pone.0079332.g001

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 4 October 2013 | Volume 8 | Issue 10 | e79332

made an attempt that lasted for at least 24 hours (both p’s <0.05). The same is true for snus use: those who are currentlyusing snus are more likely to have tried to quit smoking thanthose not currently using snus. The former are also more likelyto have made an attempt that lasted for at least 24 hours (bothp’s < 0.05).

Table 5 shows data on those reporting having “ever heard ofe-cigarettes” and where they heard about them. Three quartersof survey respondents, 75.4%, reported that they have heard ofe-cigarettes. The rate of awareness is high across gender, age,education level and ethnicity. Even 69.2% of never smokershave heard about e-cigarettes, and the percentage goes up to88.1% for current smokers.

Table 5 also shows that those who have heard about e-cigarettes are most likely to report television as their source ofinformation, 48.0%. The second most likely source is “in-person conversation”, 38.2%, followed by Internet, 20.7%, andradio, 12.2% and social networks, 2.7%.

There are some interesting differences in Table 5. Forexample, men are more likely to have heard about e-cigarettesthan women, in general. Older people are more likely thanyounger people to have heard about e-cigarettes fromtelevision, while being less likely to have heard about themfrom the Internet. The same pattern appears in the lower andhigher education groups. Smokers and recent former smokersare more likely to have heard about e-cigarettes “in-person”than long-term former smokers or never smokers, while thelatter groups are more likely to have heard about them fromtelevision. There is a clear trend that television and in-personare the most common sources of awareness for e-cigarettes.

Survey respondents can report more than one source ofawareness of e-cigarettes. However most, 72.1%, reportedonly one source when answering the survey. Another 18.4%checked off two sources. The rest, 9.5%, reported three ormore sources (data not shown in Table 5).

Table 1. Ever and Current Use of E-Cigarettes and Snus (n = 10,041).

Ever Use

E-cigarettes only Snus only E-Cigarettes and Snus % (95% CI) % (95% CI) % (95% CI)Mean 6.89 (6.40-7.39) 3.07 (2.73-3.41) 1.19 (0.98-1.40)

Gender Male 5.33 (4.51-6.15) 5.26 (4.32-6.20) 1.84 (1.25-2.43) Female 8.33 (7.33-9.33) 1.06 (0.63-1.49) 0.59 (0.28-0.90)

Age 18-24 9.33 (6.59-12.07) 3.54 (1.60-5.48) 3.37 (1.49-5.25) 25-44 7.78 (6.43-9.13) 3.98 (2.92-5.04) 1.69 (1.02-2.36) 45-54 7.35 (6.47-8.23) 2.88 (2.23-3.53) 0.52 (0.28-0.76) 65+ 2.62 (19.3-3.31) 1.25 (0.78-1.72) 0.34 (0.00-0.69)

Education ≤12 years 8.60 (7.37-9.83) 3.48 (2.60-4.36) 1.61 (1.23-1.99) >12 years 5.63 (4.94-6.32) 2.76 (2.17-3.35) 0.88 (0.61-1.15)

Ethnicity Non-Hispanic White 7.02 (6.29-7.75) 3.27 (2.66-3.88) 1.32 (0.91-1.73) Black 8.12 (5.55-10.69) 1.19 (0.39-1.99) 0.35 (0.00-0.74) Hispanic 4.38 (1.81-6.95) 4.40 (1.64-7.16) 1.42 (0.00-3.07) Other 6.14 (4.87-7.40) 2.94 (1.43-4.45) 1.22 (0.64-1.80) Multi-racial 9.51 (5.24-13.78) 4.69 (0.97-8.41) 0.52 (0.00-1.07)

Current Use

E-cigarettes only Snus only E-Cigarettes and Snus % (95% CI) % (95% CI) % (95% CI)Mean 1.28 (1.06-1.50) 0.61 (0.46-0.76) 0.16 (0.08-0.24)

Gender Male 1.00 (0.65-1.35) 0.96 (0.53-1.39) 0.22 (0.00-0.46) Female 1.54 (1.13-1.95) 0.29 (0.11-0.47) 0.10 (0.00-0.26)

Age 18-24 1.47 (0.33-2.61) 0.27 (0.00-0.58) 0.23 (0.00-0.48) 25-44 1.19 (0.62-1.76) 0.75 (0.28-1.22) 0.29 (0.00-0.62) 45-54 1.67 (1.34-2.00) 0.69 (0.30-1.08) 0.08 (0.00-0.22) 65+ 0.54 (0.25-0.83) 0.33 (0.02-0.64) 0.02 (0.00-0.04)

Education ≤12 years 1.78 (1.21-2.35) 0.81 (0.40-1.22) 0.35 (0.04-0.66) >12 years 0.91 (0.69-1.13) 0.47 (0.22-0.72) 0.02 (0.00-0.04)

Ethnicity Non-Hispanic White 1.53 (1.18-1.88) 0.59 (0.33-0.85) 0.06 (0.00-0.14) Black 1.26 (0.32-2.20) 0.49 (0.00-0.99) 0.03 (0.00-0.07) Hispanic 0.30 (0.00-0.72) 1.87 (0.00-3.97) 0.70 (0.00-2.07) Other 0.58 (0.16-1.00) 0.31 (0.00-0.63) 0.54 (0.00-1.23) Multi-racial 0.79 (0.19-1.39) 0.16 (0.00-0.47) 0

doi: 10.1371/journal.pone.0079332.t001

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 5 October 2013 | Volume 8 | Issue 10 | e79332

Table 6 presents the percentage of population who can beconsidered susceptible to future e-cigarette use. The“susceptible” category includes all those who had ever tried e-cigarettes but were not currently using, and those who have notexperimented with them but are “very likely”, or “somewhatlikely” to use them in the future. The proportion of respondentssusceptible to future use is dramatically different acrosssmoking status. About 2.6% of never smokers are susceptible,

but nearly half, 49.5%, of current smokers are susceptible tofuture use of e-cigarettes. Among those who are susceptible,an average of 56.4% have tried e-cigarettes but are notcurrently using them, 9.5% said they are “very likely” to usethem, and 34.1 % said “somewhat likely.”

Table 2. Ever and Current Use of E-Cigarettes and Snus, by Gender and Smoking Status.

Never smokers (n=3,254) Long-term former smokers*(n=3,263) Recent former smokers† (n=413) Current smokers (n=3,111)

% (95% CI) % (95% CI) % (95% CI) % (95% CI)Ever use of e-cigarettes Male 0.97 (0.34-1.60) 1.70 (0.92-2.48) 24.49 (16.94-32.04) 26.99 (23.07-30.91) Female 1.09 (0.48-1.70) 3.17 (2.05-4.29) 29.11 (21.37-36.85) 37.57 (33.55-41.59) Mean 1.04 (0.61-1.47) 2.40 (1.71-3.09) 26.78 (21.35-32.21) 32.18 (29.34-35.05)

Ever use of snus Male 2.57 (1.39-3.75) 6.37 (4.76-7.98) 22.93 (14.66-31.20) 17.27 (13.74-20.80) Female 0.50 (0.21-0.79) 0.49 (0.06-0.92) 5.77 (1.77-9.77) 6.08 (3.55-8.61) Mean 1.43 (0.86-2.00) 3.45 (2.66-4.42) 14.43 (9.65-19.21) 11.76 (9.56-13.96)

Current use of e-cigarettes Male 0.05 (0.00-0.15) 0.12 (0.00-0.26) 4.97 (1.58-8.36) 4.96 (3.07-6.87) Female 0.03 (0.00-0.09) 0.22 (0.00-0.47) 7.22 (2.79-11.65) 7.61 (5.41-9.81) Mean 0.04 (0.00-0.10) 0.17 (0.03-0.31) 6.08 (3.30-8.86) 6.26 (4.81-7.71)

Current use of snus Male 0.56 (0.26-0.85) 0.64 (0.18-1.09) 2.84 (0.18-5.45) 3.20 (2.10-4.30) Female 0.06 (0.00-0.17) 0.23 (0.00-0.50) 0.0 1.70 (0.51-2.82) Mean 0.28 (0.00-0.57) 0.44 (0.11-0.77) 1.44 (0.09-2.79) 2.46 (1.46-3.46)

* Ever smokers who quit more than a year ago at the time of survey.† Ever smokers who quit within a year or less at the time of survey.doi: 10.1371/journal.pone.0079332.t002

Table 3. Frequency of Using E-Cigarettes and Snus in the Past 30 Days.

Never smokers Long-term former smokers* Recent former smokers† Current smokers Overall % (95% CI) % (95% CI) % (95% CI) % (95% CI)

Current use of e-cigarettes (n=267) Every Day 0 31.0 (0.0-71.7) 45.7 (20.9-70.5) 11.5 (3.3-19.7) 16.3 (8.2-24.4) Some Days 100 69.0 (28.3-100.0) 54.3 (29.5-79.1) 88.5 (80.3-96.7) 83.7 (75.6-91.8)

Current use of snus (n=80) Every Day 11.3 (0.0-33.8) 75.3 (47.7-100.0) 15.9 (0.0-45.7) 23.2 (1.6-44.8) 27.1 (11.1-43.1) Some Days 88.7 (66.2-100.0) 24.7 (0.0-52.3) 84.1 (54.3-100.0) 76.8 (55.2-98.4) 72.9 (56.9-88.9)

* Ever smokers who quit more than a year ago at the time of survey.† Ever smokers who quit within a year or less at the time of survey.doi: 10.1371/journal.pone.0079332.t003

Table 4. Reasons for Having Tried E-Cigarettes and Snus*.

Used either e-cigarettes or snus Used both e-cigarettes and snus

E-Cigarettes (n=1,057) Snus (n=316) E-Cigarettes (n=122) Snus (n=122)Safer than cigarettes 49.9 (44.7-55.1) 10.8 (5.9-15.7) 58.2 (44.7-71.7) 26.2 (13.8-38.5)

Cheaper than cigarettes 30.3 (25.7-34.9) 24.6 (16.7-32.5) 36.9 (22.8-51.0) 24.8 (13.6-36.0)

Easy to use when I can't smoke 44.8 (39.7-49.9) 37.6 (29.2-46.0) 57.4 (43.7-71.1) 49.9 (36.0-63.8)

To try to quit smoking cigarettes 54.9 (49.8-60.0) 26.3 (19.0-33.6) 56.9 (43.3-70.5) 30.1 (17.7-42.5)

Just because 68.3 (63.8-72.8) 73.8 (66.6-81.0) 72.3 (57.7-85.9) 82.1 (73.3-90.9)

* The order of these options was randomized for individual respondents to minimize the order effect in response.doi: 10.1371/journal.pone.0079332.t004

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 6 October 2013 | Volume 8 | Issue 10 | e79332

Discussion

This study, based on a national probability sample, foundthat three quarters of the U.S. adult population have heardabout e-cigarettes, and approximately 8% of them haveexperimented with e-cigarettes. Among current smokers, over30% have ever used e-cigarettes. Of those who have ever triede-cigarettes, about 18% are currently using them, the sametransition rate as for snus use. However, both the ever use andthe current use rates are about twice those for snus. Moreover,those who have tried e-cigarette are twice as likely to reportusing e-cigarettes as a quitting aid than snus users are toreport using snus as a quitting aid. E-cigarette users are alsosignificantly more likely to consider e-cigarettes safer thanconventional cigarettes than snus users are to consider thesame about snus. Finally, about half of current smokers appearto be susceptible to e-cigarette use in the future.

That 75% of U.S. population reported being aware of e-cigarettes is somewhat surprising, given that the survey wasconducted before the onset of any major paid media promotionby large tobacco companies. E-cigarettes are presumablymostly of interest to current smokers, or about 20% of the U.S.population [50]. This survey found that almost 90% of current

smokers have heard about them, but even two-thirds of neversmokers reported having heard of e-cigarettes. When askedabout where they have heard about these products, televisiontops the list. Internet, which we suspected to be a major driverfor the spread of information about e-cigarettes [12,54], ranksthird on the list.

Since there is little data indicating major televisionadvertising paid for by e-cigarette manufacturers before thissurvey was conducted (in February-March, 2012), the highlevel of awareness attributed to television suggests that theproducts might have garnered considerable earned mediaattention. Earned media includes national news programs orhealth programs that discuss the pros and cons of e-cigarettes[55-58]. It also includes celebrity endorsements on popular TVtalk shows [34,35]. And it could include many local televisionand radio programs, which feed from these national newsprograms [59-61]. There is, however, no formal documentationof exactly how often this took place, and survey respondentsmight have based their report on what was most salient in theirmemory, not what was most frequent.

One reason for the media’s interest in e-cigarettes may be anovelty effect as they are relatively new products in the U.S.market. Another reason that e-cigarettes may have attracted

Figure 2. Quit attempts among current users of E-Cigarettes and Snus versus non-users. doi: 10.1371/journal.pone.0079332.g002

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 7 October 2013 | Volume 8 | Issue 10 | e79332

earned media is the intuitive appeal of the products: e-cigarettes mimic regular cigarettes in so many ways that itseems to be, simply put, a clever invention. For example, aswe were preparing the first draft of this paper, articles about e-cigarettes appeared in two highly regarded Americanpublications: National Geographic and Consumer Reports[62,63]. Neither article directly promoted e-cigarettes, but both

clearly stoked interest in the products. Both articles includedappealing pictures of e-cigarettes. The fact that “in-personconversation” was the second most frequently reported sourceof information (ahead of “Internet”) in the present study alsosuggests that many people find the products interestingenough to raise the topic with friends and colleagues.

Table 5. Awareness of E-Cigarettes.

Ever heard of e-cigarettes(n=8,045)

Heard of on radio(n=1,019)

Heard of ontelevision(n=3,806)

Heard of on theinternet (n=1,850)

Heard of in-person(n=3,178)

Heard of onsocial networks(n=199)

None of theabove (n=1,473)

% (95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) Average 75.4 (74.1-76.7) 12.2 (11.2-13.2) 48.0 (46.4-49.6) 20.7 (19.4-22.0) 38.2 (36.3-39.8) 2.7 (2.1-3.3) 18.0 (16.7-19.3)

Gender Male 78.9 (77.1-80.7) 13.8 (12.2-15.4) 49.4 (47.0-51.8) 25.2 (23.2-27.2) 34.2 (32.0-36.4) 2.9 (2.1-3.7) 17.2 (15.4-19.0) Female 72.3 (70.3-74.3) 10.5 (9.1-11.9) 46.6 (44.2-49.0) 16.3 (14.5-18.1) 42.1 (39.7-44.5) 2.6 (1.8-3.4) 18.7 (16.9-20.5)

Age 18-24 76.7 (71.8-81.6) 8.9 (5.4-12.4) 41.6 (35.1-48.1) 25.1 (19.4-30.8) 46.0 (39.3-52.7) 3.2 (1.2-5.2) 14.7 (10.4-19.0) 25-44 74.9 (72.4-77.4) 13.2 (11.2-15.2) 40.7 (37.6-43.8) 23.6 (21.1-26.1) 43.5 (40.4-46.6) 3.8 (2.6-5.0) 19.3 (16.8-21.8) 45-64 78.0 (76.2-79.8) 12.3 (10.9-13.7) 52.7 (50.3-55.1) 19.5 (17.7-21.3) 36.0 (33.8-38.2) 1.8 (1.2-2.4) 16.9 (15.1-18.7) 65+ 70.2 (67.7-72.7) 11.3 (9.1-13.5) 57.3 (54.2-60.4) 14.4 (12.4-16.4) 26.2 (23.5-28.9) 2.2 (0.8-3.6) 19.3 (16.8-21.8)

Education ≤12 years 73.1 (70.9-75.3) 11.1 (9.5-12.7) 51.8 (49.1-54.5) 17.3 (15.3-19.3) 37.0 (34.5-39.5) 2.7 (1.7-3.7) 15.6 (13.6-17.6) >12 years 77.2 (75.4-79.0) 12.9 (11.5-14.3) 45.3 (43.1-47.5) 23.2 (21.4-25.0) 39.0 (37.0-41.0) 2.7 (1.9-3.5) 19.6 (17.8-21.4)

EthnicityNon-HispanicWhite

80.6 (79.4-81.8) 12.4 (11.2-13.6) 47.4 (45.6-49.2) 20.2 (18.8-21.6) 38.5 (36.7-40.3) 2.6 (2.0-3.2) 17.4 (16.0-18.8)

Black 62.3 (57.4-67.2) 11.1 (7.8-14.4) 61.1 (55.2-67.0) 20.1 (15.4-24.8) 30.9 (25.4-36.4) 2.4 (0.8-4.0) 19.7 (14.6-24.8) Hispanic 66.2 (59.1-73.3) 17.4 (11.1-23.7) 40.1 (32.3-47.9) 29.7 (22.4-37.0) 39.5 (31.3-47.7) 4.1 (0.6-7.6) 20.0 (13.1-26.9) Other 65.5 (60.8-70.2) 8.4 (5.5-11.3) 45.1 (39.2-51.0) 19.7 (15.0-24.4) 40.3 (34.6-46.0) 2.7 (0.3-5.1) 18.0 (13.7-22.3) Multi-racial 71.6 (62.4-80.8) 18.7 (8.9-28.5) 46.6 (36.4-56.8) 29.7 (19.5-39.9) 49.1 (39.9-59.3) 6.8 (0.0-15.4) 26.8 (16.0-37.6)

SmokingStatus

Never 69.2 (67.0-71.4) 12.1 (10.5-13.7) 47.5 (44.8-50.2) 18.6 (16.4-20.8) 34.4 (31.9-36.9) 2.6 (1.6-3.6) 19.2 (17.0-21.4)

LT* former 78.7 (76.9-80.5) 11.5 (9.9-13.1) 52.8 (50.3-55.3) 19.2 (17.2-21.2) 34.4 (32.0-36.8) 2.2 (1.4-3.0) 17.3 (15.5-19.1) RT† former 85.9 (81.4-90.4) 12.1 (7.6-16.6) 40.8 (34.1-47.5) 27.8 (21.5-34.1) 49.2 (42.3-56.1) 3.9 (1.0-6.8) 14.5 (9.8-19.2) Current 88.1 (85.9-90.3) 13.1 (10.9-15.3) 45.1 (42.0-48.2) 26.1 (23.4-28.8) 49.0 (45.9-52.1) 3.4 (2.2-4.6) 16.4 (14.2-18.6)

* Long-term: Ever smokers who quit more than a year ago at the time of survey.† Recent-term: Ever smokers who quit within a year or less at the time of survey.doi: 10.1371/journal.pone.0079332.t005

Table 6. Susceptibility to Using E-Cigarettes in the Future*.

Never smokers(n=3,251)

Long-term former smokers†

(n=3,256)Recent former smokers‡

(n=385)Current smokers(n=2,882)

Overall sample(n=9,774)

% (95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI)Not susceptible 97.4 (96.6-98.2) 96.7 (95.9-97.5) 75.1 (69.6-80.6) 50.5 (47.4-53.6) 88.0 (87.1-88.9)

All Susceptible 2.6 (1.8-3.4) 3.3 (2.5-4.1) 24.9 (19.4-30.4) 49.5 (46.4-52.6) 12.0 (11.1-12.9)

All Susceptible bycategory

Tried e-cigarettes 38.7 (25.4-52.0) 67.1 (56.1-78.1) 88.2 (80.7-95.6) 55.8 (51.5-60.1) 56.4 (52.6-60.2)Very likely 14.9 (7.3-22.5) 16.0 (6.8-25.2) 1.0 (0.0-2.4) 8.8 (6.6-11.0) 9.5 (7.5-11.5)Somewhat likely 46.4 (32.7-60.1) 16.9 (9.1-24.7) 10.8 (3.9-17.7) 35.4 (31.3-39.5) 34.1 (30.5-37.7)

* Sample excludes those who reported current use of e-cigarettes.† Smokers who quit more than a year ago at the time of survey.‡ Smokers who quit within a year or less at the time of survey.doi: 10.1371/journal.pone.0079332.t006

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 8 October 2013 | Volume 8 | Issue 10 | e79332

A limitation of the present study is that members of theKnowledgePanel sample may engage in more than one surveyin a given year, which might lead to greater familiarity withcertain topics. It is possible that the rate of self-reportedawareness of e-cigarettes might have been inflated for thatreason, particularly because the survey described e-cigarettesbefore asking if respondents had heard about them. However,high rates of awareness have also been reported in othersurveys. For example, a study in England found that 62% and79% of smokers were aware of e-cigarettes in 2010 and 2012,respectively [5]. Consumer surveys in the U.S. also showedhigh awareness rate: In 2010 and 2011, awareness of e-cigarettes was 41% and 58%, respectively [37]. Although theconsumer survey in the U.S. was not based on a probabilitysample of the entire population, it did show that awarenessincreased significantly. With these consumer survey data as areference, the high level of awareness found in the presentstudy suggests that interest in e-cigarettes in the U.S.continued to grow after 2011, to 75% in 2012 among the U.Spopulation as a whole (Table 5).

This increase in awareness is supported by the increase inthe rate of ever use of e-cigarettes. The consumer surveystudy, referenced above, reported that 3.3% of respondents ina web-based survey had ever used e-cigarettes in 2010, whichincreased to 6.2% in 2011 [37]. Another smaller but population-based survey in 2010 reported an ever use rate of 1.8% [64].The present study, which was population-based and conductedin March 2012, found 8.1% reported ever having used e-cigarettes (Figure 1), 3 to 4.5 times higher the rates found in2010. In contrast, ever use of snus has not increased from2010 to 2012. The ever use rates for snus from the two 2010surveys cited were 5.4% and 5.1%, respectively [64,65]. Thepresent study found 4.3% of respondents have ever used snus.In other words, the ever use rate for snus was at least twice ashigh as that of e-cigarettes in 2010. By 2012, the rate of everuse for e-cigarettes has jumped to be twice as high as that ofsnus, and the rate of snus use has remained essentiallyunchanged.

Interestingly, this survey, based on a probability sample ofthe U.S. population, found that most current users of e-cigarettes use them on a non-daily basis. This differs fromprevious studies that recruited subjects through websites,whose samples are less likely to be representative of all e-cigarette users [14,66,67]. For example, one online survey of e-cigarette users found 81% of them were daily users [7]. Thepresent study did find, however, that e-cigarette users who arerecent former smokers are much more likely to be daily usersthan those who are still smoking regular cigarettes (Table 5).This could be an indication that some of these recent quittersare using e-cigarettes daily as a replacement for regularcigarettes.

Over 50% of those who have ever used e-cigarettes reportedtrying to quit regular cigarettes as one reason they used e-cigarettes. This is supported by data that current users of e-cigarettes are indeed more likely than non-users to have madean attempt to quit regular cigarettes in the last 12 monthspreceding the survey. A smaller proportion of ever users ofsnus reported trying to quit regular cigarettes. Like e-cigarette

users, current users of snus are more likely than non-users tohave made a quit attempt in the last 12 months.

The proportion of e-cigarette users who believe that e-cigarettes are safer than regular cigarettes is significantlyhigher than snus users who believer snus is safer thancigarettes (50% vs. 11%). This perception may be incorrect,but it may have contributed to the large increase inexperimentation with e-cigarettes from 2010 to 2012, whilesnus use has remained relatively constant.

The contrast in American smokers’ interest in e-cigarettesand snus is instructive in many ways. Snus has been anestablished and popular tobacco product in Sweden for manydecades. It came to the U.S. market six years before e-cigarettes [68]. It has the support of large U.S. tobaccocompanies [27]. E-cigarettes, on the other hand, were firstdeveloped in China in 2003 and came to the U.S. market in2007 [4]. Prior to the recent acquisition of the Blu e-cigarettecompany by Lorillard, e-cigarettes were promoted mainly bysmall producers. Yet, the use of e-cigarette products has grownfrom half that of snus in 2010 to twice that of snus by 2012.

One reason that more smokers are experimenting with e-cigarettes than with snus, however, appears to be thefollowing: e-cigarettes appeal to both men and women whilesnus appeals mainly to men. In fact, e-cigarettes appeal towomen more than men (Table 2). It is possible e-cigarettes areperceived as clean nicotine devices, which might appeal towomen more than men. The design and packaging of e-cigarettes and e-cigarette promotion that is specifically targetedto women may also have contributed to this gender difference[69]. In any case, the fact that more women than men havetried e-cigarettes deserves careful investigation. It is especiallyinteresting since men are more likely to have heard about e-cigarettes than women (Table 5). No other so-called potentiallyreduced exposure product (PREP) has attracted more womenthan men.

The most striking contrast, perhaps, is between the adoptionof e-cigarettes and the adoption of another product that is verysimilar to e-cigarettes. Premier, later called Eclipse, is almostexactly the same as an e-cigarette. It does not involvecombustion when smoked [70,71], it looks like a regularcigarette, and it lights up when smoked. However, it still usestobacco leaves. It heats the tobacco leaves to deliver nicotineto smokers. The product is reported to have cost the R.J.Reynolds tobacco company about $1 billion U.S. dollars todevelop and market test [72]. There was much discussion andpromotion when the product first came to market [71,73], but itnever quite took off [73]. In contrast, e-cigarettes appear tohave tapped into the popular imagination quickly, initiallywithout the backing of any major tobacco company.

This study shows that nearly half of current adult smokers inthe U.S. are susceptible to future use of e-cigarettes, and about25% of the recent former smokers are susceptible. In addition,3.3% of long term former smokers and even 2.6% of adultnever smokers are susceptible. While the rates for these lattertwo groups are low, the size of these two groups is about 80%of the adult U.S. population. It is not clear what proportion ofyouth is susceptible to e-cigarette use. But the number ofpotential e-cigarette users among adults is already very large.

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 9 October 2013 | Volume 8 | Issue 10 | e79332

All together, the rates in Table 6 translate to 29 million adults inthe U.S. susceptible to e-cigarette use.

The popularity of e-cigarettes, if it continues to grow, createsa dilemma for the public health community. On the one hand,e-cigarettes are a new kind of tobacco-based product that iscompletely unregulated. There are numerous brands currentlyon market, easily purchased over the Internet or even in gasstations and convenience stores [74]. The ingredients of mostbrands are not reported. Safety data are lacking. Their efficacyfor helping smokers to quit regular cigarettes is not wellestablished. Their potential negative impact on tobacco controlnorms is unknown, especially their potential to induceadolescent nonsmokers to take up tobacco-based products.Meanwhile, many smokers believe e-cigarettes are safer thanregular cigarettes. Many have used them with the hope thatthey would help them quit smoking regular cigarettes. Asubstantial proportion of smokers also find e-cigarettescheaper than regular cigarettes (Table 4), which can contributeto the popularity of the former. All of these data suggest thatsmokers in the U.S. are not waiting for a consensus view fromhealth authorities to decide if they should switch to e-cigarettes.The results of the present study and those of previous studiessuggest that e-cigarettes are likely to gain users in the next fewyears regardless of the opinions of the scientific community.

The fact that e-cigarettes have quickly surpassed snus inperceptions related to safety and utility, and in actual useamong U.S. smokers suggests that some feature of e-cigarettes must have tapped into smokers’ intuitivepreferences. Whether these beliefs are correct or not, theycould potentially be channeled into a productive public healthcampaign to increase the rate of current smokers trying to quitcigarettes. Given that the population smoking cessation ratehas not improved in the last twenty years in the U.S., anymeasure that could increase the rate of smokers attempting toquit deserves consideration [75]. The rate of current use of e-

cigarettes is still relatively low, and there has been no studysuggesting that their coming to the market has led to anydetectable change in the quit attempt rate at the populationlevel. But more research on e-cigarettes or similar productsthat have a strong intuitive appeal may help in developing aconceptual model and corresponding policy to increase thepopulation cessation rate.

The case of e-cigarettes and their rapid adoption, inconjunction with the lack of scientific data on safety andefficacy, presents a difficult regulatory problem. It is imperativethat the scientific community rise to the challenge. The usualapproach to research for any product intended to help smokersquit using regular cigarettes proceeds from safety to efficacy.Such a process usually takes many years, and millions may beusing e-cigarettes before that process is completed. Studiesare needed to assess risks and benefits of these new productsfor individual users more rapidly. Equally important, studies areneeded to identify factors that influence the population usepatterns and to determine how individual preference for variousproducts translates into benefit or harm on the population level.

Acknowledgements

We would like to thank David Cowling for his assistance inquestionnaire design and statistical analysis. We would alsolike to thank Gary Tedeschi, Yue-Lin Zhuang, and ChristopherAnderson for their comments on earlier drafts of themanuscript.

Author Contributions

Conceived and designed the experiments: SHZ. Analyzed thedata: SHZ AG LY ML. Wrote the manuscript: SHZ AG SC MLLY LZ. Survey Design: SHZ SC LZ.

References

1. Cobb NK, Byron MJ, Abrams DB, Shields PG (2010) Novel nicotinedelivery systems and public health: The rise of the "E-cigarette". Am JPublic Health 100(12): 2340-2342. doi:10.2105/AJPH.2010.199281.PubMed: 21068414.

2. Regan AK, Promoff G, Dube SR, Arrazola R (2013) Electronic nicotinedelivery systems: Adult use and awareness of the ‘e-cigarette’ in theUSA. Tob Contr 22(1): 19-23. doi:10.1136/tobaccocontrol-2011-050044.

3. Foulds J, Veldheer S, Berg A (2011) Electronic cigarettes (e-cigs):Views of aficionados and clinical/public health perspectives. Int J ClinPract 65(10): 1037-1042. doi:10.1111/j.1742-1241.2011.02751.x.PubMed: 21801287.

4. Pauly J, Li Q, Barry MB (2007) Tobacco-free electronic cigarettes andcigars deliver nicotine and generate concern. Tob Contr 16(5):357-357. doi:10.1136/tc.2006.019687. PubMed: 17897997.

5. Dockrell M, Morrison R, Bauld L, McNeill A (2013) E-cigarettes:Prevalence and attitudes in Great Britain. Nicotine Tob Res (advanceonline publication). doi:10.1093/ntr/ntt057. PubMed: 23703732.

6. Ayers JW, Ribisl KM, Brownstein JS (2011) Tracking the rise inpopularity of electronic nicotine delivery systems (electronic cigarettes)using search query surveillance. Am J Prev Med 40(4): 448-453. doi:10.1016/j.amepre.2010.12.007. PubMed: 21406279.

7. Etter JF, Bullen C (2011) Electronic cigarette: Users profile, utilization,satisfaction and perceived efficacy. Addiction 106(11): 2017-2028. doi:10.1111/j.1360-0443.2011.03505.x. PubMed: 21592253.

8. Pearson JL, Richardson A, Niaura RS, Vallone DM, Abrams DB (2012)E-cigarette awareness, use, and harm perceptions in US adults. Am J

Public Health 102(9): 1758-1766. doi:10.2105/AJPH.2011.300526.PubMed: 22813087.

9. Adkison SE, O'Connor RJ, Bansal-Travers M, Hyland A, Borland R etal. (2013) Electronic nicotine delivery systems: International tobaccocontrol four-country survey. Am J Prev Med 44(3): 207-215. doi:10.1016/j.amepre.2012.10.018. PubMed: 23415116.

10. Cahn Z, Siegel M (2011) Electronic cigarettes as a harm reductionstrategy for tobacco control: A step forward or a repeat of pastmistakes? J Public Health Policy 32(1): 16-31. doi:10.1057/jphp.2010.41. PubMed: 21150942.

11. Wagener TL, Siegel M, Borrelli B (2012) Electronic cigarettes:Achieving a balanced perspective. Addiction 107(9): 1545-1548. doi:10.1111/j.1360-0443.2012.03826.x. PubMed: 22471757.

12. McQueen A, Tower S, Sumner W (2011) Interviews with “Vapers”:Implications for future research with electronic cigarettes. Nicotine TobRes 13(9): 860-867. doi:10.1093/ntr/ntr088. PubMed: 21571692.

13. Polosa R, Caponnetto P, Morjaria JB, Papale G, Campagna D et al.(2011) Effect of an electronic nicotine delivery device (e-cigarette) onsmoking reduction and cessation: A prospective 6-month pilot study.BMC Public Health 11: 786. doi:10.1186/1471-2458-11-786. PubMed:21989407.

14. Siegel MB, Tanwar KL, Wood KS (2011) Electronic cigarettes as asmoking-cessation tool: Results from an online survey. Am J Prev Med40(4): 472-475. doi:10.1016/j.amepre.2010.12.006. PubMed:21406283.

15. Bullen C, McRobbie H, Thornley S, Glover M, Lin R et al. (2010) Effectof an electronic nicotine delivery device (e cigarette) on desire tosmoke and withdrawal, user preferences and nicotine delivery:

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 10 October 2013 | Volume 8 | Issue 10 | e79332

Randomised cross-over trial. Tob Contr 19(2): 98-103. doi:10.1136/tc.2009.031567.

16. Etter JF, Bullen C, Flouris AD, Laugesen M, Eissenberg T (2011)Electronic nicotine delivery systems: A research agenda. Tob Contr20(3): 243-248. doi:10.1136/tc.2010.042168. PubMed: 21415064.

17. Vansickel AR, Weaver MF, Eissenberg T (2012) Clinical laboratoryassessment of the abuse liability of an electronic cigarette. Addiction107(8): 1493-1500. doi:10.1111/j.1360-0443.2012.03791.x. PubMed:22229871.

18. Caponnetto P, Campagna D, Cibella F, Morjaria JB, Russo C et al.(2013) EffiCiency and Safety of an eLectronic cigAreTte (ECLAT) asTobacco Cigarettes Substitute: A Prospective 12-Month RandomizedControl Design Study. PLOS ONE 8(6): e66317. doi:10.1371/journal.pone.0066317. PubMed: 23826093.

19. Polosa R, Morjaria JB, Caponnetto P, Campagna D, Russo C et al.(2013) Effectiveness and tolerability of electronic cigarette in real-life: A24-month prospective observational study. Intern. Emerg Med(advance online publication). doi:10.1007/s11739-013-0977-z.

20. Study WHO Group on Product Regulation (2009) Report on thescientific basis of tobacco product regulation: Third report of a WHOstudy group. WHO Tech Rep S 955: 41.

21. Kirshner L (2011) D.C. circuit rules FDA cannot block e-cigaretteimports--Sottera, Inc. v. FDA. Am J Law Med 37(1): 194-201. PubMed:21614999.

22. Rigik E (2012) Lawmakers targeting electronic cigarettes. ConvenienceStore Decisions. Available: http://www.csdecisions.com/2012/04/12/lawmakers-targeting-electronic-cigarettes/. Accessed 8 August 2013.

23. Gilljam H, Galanti MR (2003) Role of snus (oral moist snuff) in smokingcessation and smoking reduction in Sweden. Addiction 98(9):1183-1189. doi:10.1046/j.1360-0443.2003.00379.x. PubMed:12930201.

24. Hatsukami DK, Lemmonds C, Tomar SL (2004) Smokeless tobaccouse: Harm reduction or induction approach? Prev Med 38(3): 309-317.doi:10.1016/j.ypmed.2003.10.006. PubMed: 14766113.

25. Foulds J, Kozlowski L (2007) Snus—what should the public-healthresponse be? Lancet 369(9578): 1976-1978. doi:10.1016/S0140-6736(07)60679-5. PubMed: 17498796.

26. Gartner CE, Hall WD, Chapman S, Freeman B (2007) Should thehealth community promote smokeless tobacco (snus) as a harmreduction measure? PLOS Med 4(7): e185. doi:10.1371/journal.pmed.0040185. PubMed: 17608560.

27. Mejia AB, Ling PM, Glantz SA (2010) Quantifying the effects ofpromoting smokeless tobacco as a harm reduction strategy in the USA.Tob Contr 19(4): 297-305. doi:10.1136/tc.2009.031427. PubMed:20581427.

28. Zhu SH, Wang JB, Hartman A, Zhuang Y, Gamst A et al. (2009)Quitting cigarettes completely or switching to smokeless tobacco: DoUS data replicate the Swedish results? Tob Contr 18(2): 82-87. doi:10.1136/tc.2008.028209. PubMed: 19168476.

29. Lund KE, Scheffels J, McNeill A (2011) The association between use ofsnus and quit rates for smoking: Results from seven Norwegian cross-sectional studies. Addiction 106(1): 162-167. doi:10.1111/j.1360-0443.2010.03122.x. PubMed: 20883459.

30. Noel JK, Rees VW, Connolly GN (2011) Electronic cigarettes: A new‘tobacco’ industry? Tob Contr 20(1): 81-81. doi:10.1136/tc.2010.038562.

31. Biener L, Bogen K (2009) Receptivity to Taboka and Camel Snus in aU.S. test market. Nicotine Tob Res 11(10): 1154-1159. doi:10.1093/ntr/ntp113. PubMed: 19564175.

32. Wackowski OA, Lewis MJ, Delnevo CD (2011) Qualitative analysis ofCamel Snus' website message board—users' product perceptions,insights and online interactions. Tob Contr 20(2): e1. doi:10.1136/tc.2010.037911.

33. Esterl M (2013) E-cigarettes fire up investors, regulators. Available:online.wsj.com/article/SB10001424127887324904004578535362153026902.html. Accessed10 August 2013.

34. Freeman B (2011) E-cigarettes aren't a solution - they're part of theproblem. Available: http://theconversation.edu.au/e-cigarettes-arent-a-solution-theyre-part-of-the-problem-423. Accessed 8 August 2013.

35. Grana RA, Glantz SA, Ling PM (2011) Electronic nicotine deliverysystems in the hands of Hollywood. Tob Contr 20(6): 425-426. doi:10.1136/tc.2011.043778. PubMed: 21659450.

36. Callaghan R (2011) E - cigarettes give anti-smoking lobby the vapours;electronic cigarettes in the film The Tourist will ignite debate. The WestAustralian (Perth) HEA: 3

37. King BA, Alam S, Promoff G, Arrazola R, Dube SR (2013) Awarenessand ever use of electronic cigarettes among U.S. adults, 2010–2011.

Nicotine Tob Res 15: 1623–1627. doi:10.1093/ntr/ntt013. PubMed:23449421.

38. DiSogra C (2010) Update: Address-based sampling nets success forKnowledgePanel® recruitment and sample representation. Available:http://www.knowledgenetworks.com/accuracy/spring2010/disogra-spring10.html. Accessed 8 August 2013.

39. Knowledge Networks (2012). nowledgePanel® design summary.Available: http://www.knowledgenetworks.com/knpanel/docs/knowledgePanel(R)-design-summary-description.pdf. Accessed: 8 Aug2013.

40. Dennis JM (2010) ummary of KnowledgePanel® design. Available:http://www.knowledgenetworks.com/knpanel/docs/KnowledgePanel(R)——Design-Summary-Description.pdf. Accessed: 8 Aug 2013

41. Dennis JM (2010) KnowledgePanel®: Processes & procedurescontributing to sample representativeness & tests for self-selectionbias. Available: http://www.knowledgenetworks.com/ganp/docs/KnowledgePanelR-Statistical-Methods-Note.pdf. Accessed: 8 August2013.

42. Yeager DS, Krosnick JA, Chang L, Javitz HS, Levendusky MS et al.(2011) Comparing the accuracy of RDD telephone surveys and internetsurveys conducted with probability and non-probability samples. PublicOpin Q 75(4): 709-747. doi:10.1093/poq/nfr020.

43. Dennis JM, Chatt C, Li R, Motta-Stanko A, Pulliam P (2005) Datacollection mode effects controlling for sample origins in a panel survey:Telephone versus internet. 2005 Annual Meeting of the AmericanAssociation for Public Opinion Research.

44. Chang L, Krosnick JA (2009) National surveys via RDD telephoneinterviewing versus the internet. Public Opin Q 73(4): 641-678. doi:10.1093/poq/nfp075.

45. Cowling DW, Modayil MV, Stevens C (2010) Assessing the relationshipbetween ad volume and awareness of a tobacco education mediacampaign. Tob Contr 19: i37-i42. doi:10.1136/tc.2009.030692.PubMed: 20382649.

46. Knowledge Networks (2012) nowledgePanel® demographic profile;February 2012. Available: http://www.knowledgenetworks.com/knpanel/docs/GfK-KnowledgePanel(R)—Demographic-Profile.pdf. Accessed: 8Aug 2013.

47. O'Hegarty M, Pederson LL, Nelson DE, Mowery P, Gable JM et al.(2006) Reactions of young adult smokers to warning labels on cigarettepackages. Am J Prev Med 30(6): 467-473. doi:10.1016/j.amepre.2006.01.018. PubMed: 16704939.

48. Song AV, Ling PM, Neilands TB, Glantz SA (2007) Smoking in moviesand increased smoking among young adults. Am J Prev Med 33(5):396-403. doi:10.1016/j.amepre.2007.07.026. PubMed: 17950405.

49. Pierce JP, Distefan JM, Kaplan RM, Gilpin EA (2005) The role ofcuriosity in smoking initiation. Addict Behav 30(4): 685-696. doi:10.1016/j.addbeh.2004.08.014. PubMed: 15833574.

50. Centers for Disease Control and Prevention (2011) Vital signs: Currentcigarette smoking among adults aged ≥18 years --- United States,2005--2010. MMWR Morb Mortal Wkly Rep 60(35): 1207-1212.PubMed: 21900875.

51. Agresti A, Coull BA (1998) Approximate is better than "exact" forinterval estimation of binomial proportions. Am Stat 52(2): 119-126. doi:10.1080/00031305.1998.10480550.

52. Brown LD, Cai TT, DasGupta A (2001) Interval estimation for abinomial proportion. Stat Sci 16(2): 101-117. doi:10.1214/ss/1009213286.

53. The RProject for Statistical Computing An introduction to R. Available:http://cran.r-project.org/doc/manuals/R-intro.html. Accessed: 12 August2013.

54. Myslín M, Zhu SH, Chapman W, Conway M (2013) Using Twitter toexamine smoking behavior and perception of emerging tobaccoproducts. J Med Internet Res 15(8): e174. doi:10.2196/jmir.2534.PubMed: 23989137.

55. The Doctors TVShow (2009). how synopsis - top 10 health trends of2009. Available: http://thedoctorstv.com/main/show_synopsis/90?section=synopsis. Accessed: 8 August 2013.

56. The original ePuffers on The Doctors TV show (2010). Available: http://www.youtube.com/watch?v=xBCQJldHX8U. Accessed: 8 August 2013.

57. Show Rachael Ray Going too far? medical edition. Available: http://www.rachaelrayshow.com/show/segments/view/going-too-far-medical-edition. Accessed: 8 August 2013.

58. Dr. Sanjay Gupta(CNN) TV Show on electronic cigarette. Available:http://www.youtube.com/watch?v=04rj9Tl2CV8. Accessed: 8 August2013.

59. Electronic cigarette healthy or not? Available: http://www.youtube.com/watch?v=6FtvZCIp1oY. Accessed: 8 Aug 2013

60. CBS (2012) BS NEWS report on electronic cigarettes. Available: http://www.youtube.com/watch?v=yUpbsh0OPG0. Accessed: 8 August 2013.

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 11 October 2013 | Volume 8 | Issue 10 | e79332

61. KTLA News (2012) Electronic cigarettes ignite major health debate.Available: http://www.ktla.com/news/landing/ktla-electronic-cigarettes-debate,0,1958505.story. Accessed: 8 Aug 2013

62. Consumer Reports (2012) Do e-cigarettes help smokers quit?Consumer Reports (May 2012). Available: http://www.consumerreports.org/cro/magazine/2012/04/do-e-cigarettes-help-smokers-quit/index.htm. Accessed: 8 August 2013.

63. Tellis A, Howard BC (2012) Cigarettes vs. e-cigarettes: Which is lessenvironmentally harmful? National Geographic Newswatch. Available:http://newswatch.nationalgeographic.com/2012/04/11/cigarettes-vs-e-cigarettes-which-is-less-environmentally-harmful/. Accessed: 8 August2013.

64. McMillen R, Maduka J, Winickoff J (2012). Use of emerging tobaccoproducts in the United States. J Environ Public Health 2012: 989474.doi:10.1155/2012/989474.

65. Regan AK, Dube SR, Arrazola R (2012) Smokeless and flavoredtobacco products in the U.S.: 2009 styles survey results. Am J PrevMed 42(1): 29-36. doi:10.1016/j.amepre.2011.08.019. PubMed:22176843.

66. Etter JF (2010) Electronic cigarettes: A survey of users. BMC PublicHealth 10: 231. doi:10.1186/1471-2458-10-231. PubMed: 20441579.

67. Heavner K, Duworth J, Bergen P, Nissen C, Phillips CV (2009)Electronic cigarettes (e-cigarettes) as potential tobacco harm reductionproducts: Results of an online survey of e-cigarette users. TobaccoHarm Reduction Working Paper 011 Available: http://

www.tobaccoharmreduction.org/wpapers/011.htm. Accessed: 8 August2013.

68. Henningfield JE, Fagerstrom KO (2001) Swedish match company,Swedish snus and public health: A harm reduction experiment inprogress? Tob Contr 10(3): 253-257. doi:10.1136/tc.10.3.253. PubMed:11544390.

69. V2 Cigs (2012) apor; Couture. Available: http://v2-cigs.org/vapor-couture/. Accessed: 8 August 2013.

70. Slade J, Connolly GN, Lymperis D (2002) Eclipse: Does it live up to itshealth claims? Tob Contr 11(suppl 2): ii64-ii70. PubMed: 12034985.

71. Anderson SJ, Ling PM (2008) “And they told two friends…and so on”:RJ Reynolds’ viral marketing of Eclipse and its potential to mislead thepublic. Tob Contr 17(4): 222-229. doi:10.1136/tc.2007.024273.

72. Parker-Pope T (2001) "Safer" cigarettes: A history. Available: http://www.pbs.org/wgbh/nova/body/safer-cigarettes-history.html. Accessed:8 August 2013.

73. Hickman N, Klonoff EA, Landrine H, Kashima K, Parekh B et al. (2004)Preliminary investigation of the advertising and availability of PREPs,the new “Safe” tobacco products. J Behav Med 27(4): 413-424. doi:10.1023/B:JOBM.0000042413.69425.aa. PubMed: 15559736.

74. Lee M, Zhu S, Huang Y, Mayoral A, Conway MA (2013) A survey ofmore than 250 E-cigarette brands on the Internet. 19th Annual Meetingof the Society for Research on Nicotine and Tobacco.

75. Zhu SH, Lee M, Zhuang YL, Gamst A, Wolfson T (2012) Interventionsto increase smoking cessation at the population level: How muchprogress has been made in the last two decades? Tob Contr 21(2):110-118. doi:10.1136/tobaccocontrol-2011-050371. PubMed:22345233.

Use and Perception of E-Cigarettes and Snus

PLOS ONE | www.plosone.org 12 October 2013 | Volume 8 | Issue 10 | e79332